Paleoclimatic data from 74KL and Guliya cores: New insights Govindan Rangarajan

advertisement
GEOPHYSICAL RESEARCH LETTERS, VOL. 27, NO. 6, PAGES 787-790, MARCH 15, 2000
Paleoclimatic data from 74KL and Guliya cores:
New insights
Govindan Rangarajan1
Department of Mathematics and Center for Theoretical Studies, Indian Institute of Science, Bangalore,
India
Dhananjay A. Sant
Department of Geology, M. S. University of Baroda, Vadodara, India
Abstract. We analyze multi-proxy climatic records from
the 74KL marine core and compare it with the δ 18 O record
from the Guliya ice core. These records have important implications for the evolution of the Southwest Monsoon System. We observe three distinct climatic events (viz. 19 ka
to 13.5 ka, 13.5 ka to 10 ka and 3.4 ka to present) from the
last glacial phase to the present. Even though the first two
events are well documented in the literature, our combined
analysis of data from both the cores suggests alternative
mechanisms which could have complemented and strengthened these events. The third, most recent event (occurring
between 3.4 ka to present) does not seem to have been well
documented earlier. We propose a possible mechanism to explain this event. We identify representative elements which
better capture the above events as compared to more commonly used elements.
Introduction
The multi proxy records from sediment cores / ice cores
have preserved imprints of past climatic history [Sirocko et.
al., 1993, 1996; Kyte et. al., 1993; Clemens et. al., 1996;
Thompson et. al., 1997]. Abrupt changes in the multi-proxy
data suggest changes in the climate dynamics [Kyte et. al.,
1993; Clemens et. al., 1996; Overpeck et. al., 1996; German
and Elderfield, 1990; Murray et. al., 1992a]. In this paper,
we analyze available records [Sirocko et. al., 1993, 1996;
Thompson et. al., 1997] from two cores (one marine and
another ice) starting with the last glacial epoch and ending with the present interglacial period (Holocene). These
records have important implications for the evolution of one
of the major climatic system on the globe - the Southwest
Monsoon System.
The marine core 74KL [Sirocko et. al., 1993, 1996] is
situated in western Arabian sea. We found 74KL data most
suitable for our analysis as it incorporates concentrations of
fifty-three elements such as δ 18 O, δ 13 C, Corg , quartz in silt,
dolomite and opal fraction, along with major trace and rare
earth elements. This large number allows us to study the
same climate dynamics using several elemental records.
1 Also a honorary faculty member at the Jawaharlal Nehru
Centre for Advanced Scientific Research, Bangalore, India.
Copyright 2000 by the American Geophysical Union.
Paper number 1999GL011061.
0094-8276/00/1999GL011061$05.00
The ice core data [Thompson et. al., 1997] is from
the Guliya ice cap on the Qinghai-Tibetan Plateau. The
δ 18 O record from this core is a good proxy for past atmospheric temperature over northern Tibet [Thompson et. al.,
1997]. This temperature pattern is the key driving force of
the southwest monsoon, which in turn affects the sediment
records in the 74KL core. Therefore, in this paper, we analyze the 74KL records keeping the Guliya δ 18 O record as an
important controlling parameter.
Representative elements
We now use interelemental correlation analysis of the
fifty-three element records in the core 74KL to pick representative elements from each major grouping of elements.
In the lithogenic group, we choose the total rare earth element (TREE) values as representative of the group. TREE
value is predominantly associated with detrital fraction of
the sediment which in turn represents the bulk composition
and stands as proxy for the climate over the source area
[Chaudhuri and Cullers, 1979; Taylor and McLennan, 1985;
Murray et. al., 1991]. The choice of TREE as the representative record is justified in that trace elements belonging
to the lithogenic group such as Cs, Al, Nb, Hf, Ga, K, Pb
are strongly correlated with TREE at a 1% level of significance in each of the three climatic phases (glacial, glacialinterglacial transition and Holocene).
In the biogenic group, we pick Corg as the representative
element. Corg value represents the biotic activity in the upwelling area. We find that N and Ba correlate at a 1% level
of significance with Corg in all three climatic phases. Surprisingly, quartz, dolomite, δ 18 O, opal and δ 13 C do not correlate with any of the parameters during glacial and Holocene
phases. This suggests that they could be influenced by complex dynamics.
Analysis of climatic events from the last
glacial phase to present
Next we analyze elemental concentration versus depth
profiles and look for major events in the time series data
starting from the glacial phase and progressing to the present.
The well-documented transition between glacial and Holocene comprises two major events – the first from 19 ka and
13.5 ka and the second from 13.5 ka to 10 ka. But these two
events are not captured distinctly by the commonly used
δ 18 O, quartz, opal and δ 13 C records from the 74KL core.
These events, however, are clearly seen in all other element
787
788
RANGARAJAN AND SANT: NEW INSIGHTS FROM PALEOCLIMATIC DATA
1.4
event
2
event
1
Corg
1.0
0.6
event
3
0.2
0
5
10
15
Age (ka)
20
25
The 74KL Corg record, a representative component from biogenic group. Data obtained from ftp://ftp.ngdc.
noaa.gov, in the directory /paleo/paleocean/by contributor/
sirocko1993.
Figure 1.
records, which correlate well with the biogenic and lithogenic
flux.
cause the continent had undergone a long arid phase during glacial time whereby a large influx of eolian sediments
would have already been blown into the sea. But during
the deglaciation phase, we observe a sudden additional increase in the influx of sediments and a change in composition. This indicates that an additional process came into
play at this time. This process could very well have been the
high fluvial influx from the Himalayas that we have proposed
above. The unstable conditions resulting from this large influx would also contribute to decrease in biotic activity and
inorganic precipitation of carbonates and dolomite.
From 16.5 ka to 13.5 ka (the second half of the first event),
values for the biogenic group (Corg , N, Ba) increase by 139%
on the average, those for trace elements decrease by 39%
on the average and finally, values for rare earth elements
decrease by 36% on the average. The accepted explanation
[Sirocko et. al., 1993, 1996] for this change is the increase in
southwest monsoon intensity leading to increased upwelling
(and therefore increased biotic activity). From the Guliya
δ 18 O record (Figure 3), we notice that δ 18 O has increased
to its maximum value during this period indicating high
temperatures over Tibet. This is consistent with the above
explanation.
First event (19 ka to 13.5 ka)
Second event (13.5 ka to 10 ka)
Next, we describe the second event, which lasted approximately between 13.5 ka and 10 ka. This event starts with
a sharp decrease of 53% in the value of Corg , 40% drop in
N and 19% drop in Ba values. On the other hand, trace
and rare earth elements increase by an average of 20% and
13% respectively. The above observations can be explained
as follows. A sharp decrease in the δ 18 O value starting at
13.5 ka has been previously observed [Thompson et. al.,
1997] in the Guliya core data corresponding to the start of
the Younger Dryas. This decrease is correlated to a corresponding decrease in temperature over Tibet resulting in
weaker southwest monsoons and a more arid phase. These
conditions lead to decreased biotic activity and increased
lithogenic influx. The above event ends with an increase in
biogenic components and a decrease in trace and rare earth
components reversing the above trend. This has been well
documented [Sirocko et. al., 1993, 1996].
60
event
3
event
2
event
1
50
TREE
We now analyze the first of the two events mentioned
above in greater detail. Plots of representative element
records from each major group viz. Corg for the biogenic
group and TREE for the lithogenic group are given in Figures 1 and 2. From these figures, we observe a stable climatic
regime in the late glacial period up to around 19 ka. The
first event starts at the end of this period. The Corg , N
and Ba values drop sharply up to 16.5 ka. The value drops
by 73% for Corg (58% and 53% for N and Ba respectively)
as compared to the mean value during the preceding stable
regime. Values for trace and rare earth elements, however,
increase by an average of 14% and 8% respectively.
We now explain the above observations. An earlier work
[Sirocko et. al., 1993, 1996] had interpreted the change in
elemental values during this period as being caused by decrease in upwelling productivity and large eolian dust influx
from an arid Arabia. We give below alternative mechanisms
which could have complemented and strengthened the above
mechanism. We now have available δ 18 O record from the
Guliya core [see Figure 3]. This suggests that deglaciation
was in full swing in the western Kunlun Shan during the
above period as evidenced by a sharp increase in δ 18 O values. Thus, there was large-scale melting of ice sheets resulting in a sudden large-scale influx of water into the Arabian
Sea. This would have contributed to the observed increase
in trace and rare earth element values for reasons given below. As Himalayas is one of the youngest, active mountain
chains, tectonic events would have lead to rapid mechanical breakdown of rocks thus exposing primary minerals for
chemical weathering [Raymo et. al., 1988]. It is supposed
that dissolved loads are highest in areas dominated by earlyeroded sedimentary rocks. Further, during glacial time (110
ka to 19 ka), sea levels were low and large continental areas were exposed [Fairbanks, 1989]. Therefore, rivers flowing
from Himalayas during this time (in particular, Indus) must
have flown further deep into Indian Ocean, affecting the sedimentation at 74KL site.
Further, the eolian dust influx from Arabia could not
have solely contributed to the sudden increase. This is be-
40
30
0
5
10
15
Age (ka)
20
25
Figure 2. The 74KL TREE record, a representative component
from lithogenic group. Same source of data as in Figure 1.
RANGARAJAN AND SANT: NEW INSIGHTS FROM PALEOCLIMATIC DATA
-5
event
3
event
2
event
1
δ18O
-10
-15
-20
-25
0
5
10
15
Age (ka)
20
25
Figure 3. The Guliya δ18 O record. Data obtained from Prof.
L. G. Thompson.
Third event (3.4 ka to present)
Finally, we start our analysis of the Holocene period.
During early and middle Holocene, the climate stabilized
reflecting a moist phase. In this period, records of all elements show a stable phase with oscillations around mean
value. Around 3.4 ka, we observe a significant change in
most of the 74KL records other than δ 18 O, quartz, opal,
δ 13 C and Cd records. We performed a statistical analysis
to confirm the presence of this event by first dividing the
Holocene into two periods – 10 ka to 3.4 ka (early and middle Holocene) and 3.4 ka to present (late Holocene). The
mean values for these two periods were compared using the
standard t-test (from statistics) for each of the records. For
all records other than those listed above, a statistically significant change in the mean value (at the 1% level) was
observed. To our knowledge, the above event has not been
well documented in the literature.
From the above statistical analysis, we find that the biogenic components (Corg , N and Ba) increased significantly
during late Holocene by an average of 36%. Surprisingly,
the lithogenic components (trace and rare earth elements)
also increased significantly in this period (by an average of
20% and 27% respectively). This is a departure from the
pattern observed during the transition period.
We now give a possible explanation for the above observations. During the long, moist phase of early and middle
Holocene, the continental rocks would have undergone deep
weathering generating collouvium over the host rocks. From
the Guliya δ 18 O record, we observe a general increase in
δ 18 O value starting from 3.2 ka onwards. This corresponds
to an increase in temperature over Tibet leading to melting
of ice caps at higher altitudes. The snow melt could have
easily carried the readily available collouvium downstream
to Arabian sea leading to increase in values of trace and
rare earth elements in the 74KL core sample. The Southwest monsoon precipitation would have added to this effect.
Studies have shown dust flux is just 5% of the fluvial flux
rate which is about 1.5 - 2.0 × 1016 gm / year [Hovan and
Rea, 1992]. Hence, fluvial influx is a plausible mechanism for
transport of trace and rare earth elements. To summarize,
weathering took place during the early and middle Holocene
whereas the actual erosion of the surface started around 3.2
ka. The steady inflow hypothesized above (as opposed to
large scale deglaciation) would have also lead to an increase
in upwelling and an increase in biogenic components.
789
On the other hand, using pollen data [Singh et. al., 1990]
from lakes in Rajasthan, India, the late Holocene period
has been inferred to be more arid as compared to the midHolocene period. Nevertheless, this does not imply aridity
over the entire subcontinent (this inhomogeneity in precipitation pattern can be observed even in present day). This
is supported by archeological evidences from Indus valley
civilization (3100-1900 year B.C) which suggest numerous
flood events [Rajaram, 1999; Radhakrishnan, 1999; Lambrick, 1967; Raikes, 1964; Dales, 1966]. According to them,
local aridity in the Kutch - Rajasthan - Indus Plain region
had compelled the population to migrate to the Indus valley. Further, it was the great Indus flood that played an
important role in the decline of the Indus civilization.
Moreover, records from bog, lake and fluvial sediments
in the Spiti Valley in the Tethys Himalayas [Wake and
Mayewski, 1995] and a record from Goa [Sarkar et. al., 1999]
indicate moist phases in late Holocene. Hence, local factors
can explain the apparent contradictions in climatic records
from different sites. Further, inflow due to precipitation is
hypothesized to constitute only a secondary, supplementing
factor in the fluvial influx.
Acknowledgments. Data on 74KL core was obtained
from the National Geophysical Data Center data bank, ftp://ftp.
ngdc.noaa.gov, in the directory /paleo/paleocean/by contributor/
sirocko1993. We thank Prof. L. G. Thompson for giving us the
data from the Guliya ice core used in this paper and for his comments. We thank Prof. J. T. Overpeck for his comments. DAS
was supported by Indian Institute of Science during his visit there
when this work was carried out.
References
Chaudhuri, S. and R. L. Cullers, The distribution of rare earth
elements in deeply buried Gulf coast sediments, Chem. Geol.,
24, 327-338, 1979.
Clemens, S. C., D. W. Murray, and W. L. Prell, Nonstationary
phase of the Plio-Pleistocene Asian monsoon, Science, 274,
943-948, 1996.
Dales, G. F., The decline of the Harappans, Scientific American,
241, 92-100, 1966.
Fairbanks, R. G., A 17,000 years glacio-eustatic sea level record:
influence of glacial melting rates on the younger Dryas event
and deep ocean circulation, Nature, 342, 637-642, 1989.
German, C. R. and H. Elderfield, Rare earth elements in NW
Indian Ocean, Geochimica et Cosmochimica Acta, 54, 19291940 (1990).
Hovan, S. A., and D. K. Rea, The Cenozoic record of continental mineral deposition on broken 90E ridge Indian Ocean:
S. Africa aridity and sediment delivery from the Himalayas,
Palaeooceonography, 7, 8333-8360, 1992.
Kyte, F. T., M. Leinen, G. R. Heath, and L. Zhou, Cenozoic sedimentation history of the Central North Pacific: Inference from
the elemental geochemistry of core LL44-GPC3, Geochimica
et Cosmochimica Acta, 57, 1719-1740, 1993.
Lambrick, H. T., The Indus flood plain and the “Indus Civilization”, Geographical Journal, 133, 483-495, 1967.
Murray, R. W., et al., Rare earth, major and trace elements in
chert from of the Franciscan Complex and Monterey Group,
California: Assessing REE sources to fine-grained sediments,
Geochimica et Cosmochimica Acta, 55, 1875-1895, 1991.
Murray, R. W., et al., Interoceanic variation in the rare earth,
major and trace element deposition chemistry of chert: Perspectives gained from the DSDP and ODP records, Geochimica
et Cosmochimica Acta, 56, 1897-1913, 1992.
Overpeck, J., D. Anderson, S. Trumbore, and W. Prell, The
southwest Indian monsoon over the last 18000 years, Climate
Dynamics, 12, 213-225, 1996.
Radhakrishnan, B. P., Holocene chronology and Indian prehistory, Geol. Soc. Ind. Memoir, 42, xvii-xxv, 1999.
790
RANGARAJAN AND SANT: NEW INSIGHTS FROM PALEOCLIMATIC DATA
Rajaram, N. S., Sarasvati civilization in the Harappan seals, Geol.
Sco. Ind. Memoir, 42, 63-69, 1999.
Raikes, R. L., The end of the ancient cities of the Indus, American
Anthropologist, 66, 284-299, 1964.
Raymo, M. E., W. F. Ruddiman, and P. N. Froelich, Influence of
late Cenozoic mountain building on ocean geochemical cycles,
Geology, 16, 649-653, 1988.
Sarkar, A., R. Ramesh, B. L. K. Somayajulu, R. Agnihotri, A. J.
T. Jull, and G. S. Burr, High resolution Holocene palaeomonsoon record from Eastern Arabian Sea, Earth and Planetary
Science Letters, submitted, 1999.
Singh, G., R. J. Wasson, and D. P. Aggarwal, Vegetational and
seasonal climatic changes since the last full glacial in the Thar
Desert, northwestern India, Review of Palaeobotany and Palynology, 64, 351-358, 1990.
Sirocko, F., M. Sarnthein, H. Erienkeuser, H. Lange, M. Arnold,
and J. C. Duplessy, Century-scale events in monsoonal climate
over past 24,000 years, Nature, 354, 322-324, 1993.
Sirocko, F., D. Grabe-Schonberg, A. McIntyre, and B. Molfino,
Teleconnection between subtropical monsoon and highlatitude climates during the last glaciation, Science, 272, 526529, 1996.
Taylor, S. R. and S. M. McLennan, The continental crust: Its
composition and evolution, Blackwell, 1985.
Thompson, L. G., T. Yao, M. E. Davis, K. A. Henderson, E. M.
Thompson, P. -N. Lin, J. Beer, H. -A. Synal, J. Cole-Dai, and
J. F. Bolzan, Tropical climate instability: The last glacial cycle
from Qinghai-Tibetan ice core, Science, 276, 1821-1825, 1997.
Wake, C. P. and P. A. Mayewski (Eds.), Summary of existing
paleoclimatic records from the highlands of Central Asia, in
Report on the International Himalayan/ Tibetan Plateau Paleoclimate Workshop, Kathmandu, 1995.
G.Rangarajan, Department of Mathematics and Center for
Theoretical Studies, Indian Institute of Science, Bangalore 560
012, India. (e-mail: rangaraj@math.iisc. ernet.in)
D. A. Sant, Department of Geology, M. S. University of Baroda,
Vadodara 390 002, India.
(Received September 09, 1999; accepted January 20, 2000.)
Download